[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAE-0n50q5GJ_q9Pojjrh+34W+i6BXhxSDCS6M_6NLP0pyZYuQg@mail.gmail.com>
Date: Fri, 10 Jun 2022 13:38:21 -0700
From: Stephen Boyd <swboyd@...omium.org>
To: Dikshita Agarwal <quic_dikshita@...cinc.com>,
linux-kernel@...r.kernel.org, linux-media@...r.kernel.org
Cc: linux-arm-msm@...r.kernel.org, stanimir.varbanov@...aro.org,
quic_vgarodia@...cinc.com
Subject: Re: [PATCH] venus: Add support for SSR trigger using fault injection
Quoting Dikshita Agarwal (2022-06-09 04:52:46)
> diff --git a/drivers/media/platform/qcom/venus/dbgfs.c b/drivers/media/platform/qcom/venus/dbgfs.c
> index 52de47f..a0bfb9e 100644
> --- a/drivers/media/platform/qcom/venus/dbgfs.c
> +++ b/drivers/media/platform/qcom/venus/dbgfs.c
> @@ -4,13 +4,31 @@
> */
>
> #include <linux/debugfs.h>
> +#include <linux/fault-inject.h>
>
> #include "core.h"
>
> +#ifdef CONFIG_FAULT_INJECTION
> +static DECLARE_FAULT_ATTR(venus_ssr_attr);
> +#endif
> +
> +bool venus_fault_inject_ssr(void)
> +{
> +#ifdef CONFIG_FAULT_INJECTION
> + return should_fail(&venus_ssr_attr, 1);
> +#else
> + return false;
> +#endif
> +}
It would be better to remove the ifdef in this function and define a
static inline version that returns false when CONFIG_FAULT_INJECTION is
disabled so that the compiler doesn't have to insert a function call to
venus_fault_inject_ssr() when the config is disabled. It may also be
good to avoid the jump when enabled by exporting the attribute to the
irq handler file.
> +
> void venus_dbgfs_init(struct venus_core *core)
> {
> core->root = debugfs_create_dir("venus", NULL);
> debugfs_create_x32("fw_level", 0644, core->root, &venus_fw_debug);
> +
> +#ifdef CONFIG_FAULT_INJECTION
> + fault_create_debugfs_attr("fail_ssr", core->root, &venus_ssr_attr);
> +#endif
> }
>
> void venus_dbgfs_deinit(struct venus_core *core)
Powered by blists - more mailing lists